This proposal is focused on the structure and function of retroviral and cellular proteins encoded by the myc, myb, and erbA oncogenes. These nuclear oncoproteins appear to act as mediators of the final steps in signal transduction pathways, conveying information from the cytoplasm to the nucleus, and acting to regulate proliferation and differentiation. Recent research has resulted in the identification and characterization of these nuclear oncoproteins, and has uncovered considerable complexity among the translation products, including alternative translational initiations, post-translational modifications, and specific interactions with other macromolecules. Such complexity may be generated in the cytoplasm and act to influence function in the nucleus. The aim of this proposal is to define both how this complexity is determined and how it affects function. The Myc oncoproteins have two alternative translational initiation condons: the first AUG (located in exon 2) and an upstream in-frame CUG in exon 1. These act to produce two Myc proteins with distinct N-termini. In many of the wide variety of human and animal tumors where the c-myc gene is rearranged the translation of the CUG initiated protein is repressed due to loss or mutation of exon 1. Using retroviral vectors engineered to express either Myc protein, we propose to determine whether there are functional differences between the two protein forms and whether suppression of the CUG initiated protein contributes to oncogenic activation of Myc. We will also explore the idea that translational initiation factors play a role in the unusual initiation. Experiments designed to examine how, and in what manner, the two Myc protein forms interact with each other and with cellular macromolecules are also proposed. In addition both the Myc and Myb nuclear oncoproteins are phosphorylated in vivo by casein kinase II (CKII), a ubiquitous growth factor-activated kinase. Having defined the sites of CKII phosphorylation we propose to determine whether mutation of these sites leads to altered function or oncogenicity of Myc and Myb. The c-erb A oncogene encodes a thyroid hormone receptor. In the cell multiple c-ErbA proteins are detected. These comprise the full length c-ErbA and a C-terminally nested set of smaller proteins which contain the ligand binding domain, lack DNA binding regions, and are less tightly bound to the nucleus. We plan to determine whether these smaller ErbA proteins are derived by internal translational initiations and whether they can modulate ErbA's ability to transcriptionally activate specific genes.